CtrlK
BlogDocsLog inGet started
Tessl Logo

deepgram-go-voice-agent

Use when writing or reviewing Go code in this repo that runs a Deepgram Voice Agent session over WebSockets, including runtime settings, prompt updates, speak updates, injected messages, and event handling. Route standalone STT/TTS work to deepgram-go-speech-to-text or deepgram-go-text-to-speech.

70

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable SDK skill: executable quick start, concrete API surface, pointed gotchas, and cleanly layered external references. The main gaps are the unshown streaming/event-loop body in the example and slightly thin explicit validation checkpoints in the runtime workflow.

Suggestions

Flesh out the quick-start run() with a minimal concurrent audio-stream + event-loop skeleton (even a select on the channel handler) so the example is fully executable end-to-end.

Add an explicit validation checkpoint after Connect() — e.g., wait for the Welcome event / check the WelcomeResponse before Start() and streaming — to give the runtime workflow a concrete feedback loop.

Trim or condense the 'Central product skills' section so it reads as navigation rather than promotion, keeping the body lean.

DimensionReasoningScore

Conciseness

The body is sectioned and lean — quick start, key parameters, layered API reference, gotchas — without explaining concepts Claude already knows; the closing 'Central product skills' plug is the main trimmable bit, keeping it just below the lean-5 anchor.

4 / 5

Actionability

A copy-paste-ready Go quick start plus concrete constructors, runtime methods, and payload file paths give mostly executable guidance; the actual audio-streaming/event-loop body is left as a comment rather than shown, a minor gap below 5.

4 / 5

Workflow Clarity

A clear setup sequence (settings → handler → connect → start → defer Stop) is present and the gotchas flag bool-returning Connect(); this is a live runtime rather than destructive/batch flow, so no validation cap applies, but explicit checkpoints are thin versus the 5 anchor.

4 / 5

Progressive Disclosure

No bundle files ship with the skill, but the body is a clear overview with well-signaled, one-level-deep outward references (in-repo source files, OpenAPI/AsyncAPI/Context7, product docs, example paths) and easy section navigation, matching the 5 anchor.

5 / 5

Total

17

/

20

Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description with an explicit 'Use when…' trigger, a concrete multi-action what-list, and deliberate routing away from sibling STT/TTS skills. The only minor gap is trigger-term variety, which leans on technical jargon over user-natural synonyms.

DimensionReasoningScore

Specificity

Lists five concrete actions — 'runtime settings, prompt updates, speak updates, injected messages, and event handling' — over a named Voice Agent WebSocket domain, matching the comprehensive-coverage anchor.

5 / 5

Completeness

An explicit 'Use when writing or reviewing Go code… that runs a Deepgram Voice Agent session over WebSockets' trigger pairs with a concrete what-list, clearly answering both what and when.

5 / 5

Trigger Term Quality

Natural terms like 'Voice Agent', 'Go code', 'WebSockets', and 'STT/TTS' are present, but coverage is technical-jargon heavy and omits synonyms/file extensions, fitting the 'good coverage, a few natural terms missing' anchor rather than 5.

4 / 5

Distinctiveness Conflict Risk

It carves a clear Voice Agent runtime niche and explicitly routes standalone STT/TTS to sibling skills, minimizing overlap risk.

5 / 5

Total

19

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
deepgram/deepgram-go-sdk
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.